Dynamic multi-robot coverage framework via A*-optimized region patrolling and localized re-planning

This paper discusses the problem of building efficient coverage paths for a team of robots. An efficient multirobot coverage algorithm should result in a coverage path for every robot, such that the union of all paths generates a full coverage of the terrain and the total coverage time is minimized. A method, underlying several coverage algorithms, suggests the use of spanning trees as base for creating coverage paths. Current studies assume that the spanning tree is given, and try to make the most out of the given configuration. However, overall performance of the coverage is heavily dependent on the given spanning tree. This paper tackles the open challenge of constructing a coverage spanning tree that minimizes the time to complete coverage. We argue that the choice of the initial spanning tree has far reaching consequences concerning the coverage time, and if the tree is constructed appropriately, it could considerably reduce the coverage time of the terrain. Therefore the problem studied here is finding spanning trees that would decrease the coverage time of the terrain when used as base for multi-robot coverage algorithms. The main contributions of this paper are twofold. First, it provides initial sound discussion and results concerning the construction of the tree as a crucial base for any efficient coverage algorithm. Second, it describes a polynomial-time tree construction algorithm that, as shown in extensive simulations, dramatically improves the coverage time even when used as a basis for a simple, inefficient, coverage algorithm

The problem of distributed area coverage using multiple mobile robots is an important problem in distributed multi-robot sytems. Multi-robot coverage is encountered in many real world applications, including unmanned search & rescue, aerial reconnaissance, robotic demining, inspection of engineering structures, and automatic lawn mowing. To achieve optimal coverage, robots should move in an efficient manner and reduce repeated coverage of the same region that optimizes a certain performance metric such as the amount of time or energy expended by the robots. This dissertation especially focuses on using mini-robots with limited capabilities, such as low speed of the CPU and limited storage of the memory, to fulfill the efficient area coverage task. Previous research on distributed area coverage use offline or online path planning algorithms to address this problem. Some of the existing approaches use behavior-based algorithms where each robot implements simple rules and the interaction between robots manifests in the global objective of overall coverage of the environment. Our work extends this line of research using an emergent, swarming based technique where robots use partial coverage histories from themselves as well as other robots in their vicinity to make local decisions that attempt to ensure overall efficient area coverage. We have then extended this technique in two directions. First, we have integreated the individual-robot, swarming-based technique for area coverage to teams of robots that move in formation to perform area coverage more efficiently than robots that move individually. Then we have used a team formation technique from coalition game theory, called Weighted Voting Game (WVG) to handle situations where a team moving in formation while performing area coverage has to dynamically reconfigure into sub-teams or merge with other teams, to continue the area coverage efficiently. We have validated our techniques by testing them on accurate models of e-puck robots in the Webots robot simulation platform, as well as on physical e-puck robots. Keywords: Multi-robot Area Coverage, Weighted Voting Games, Multi-robot Coalition Formation

This study presents a motion planning method for coordinating multiple mobile robots for collision‐free navigation. A control architecture that uses cloud computing is proposed for the acquisition of real‐time robotic data and to coordinate multiple robots in an unstructured environment. The system consists of collision‐free path planning for multiple robots, obstacle avoidance, and navigation control. The path planning method was developed based on the optimal reciprocal collision avoidance (ORCA) algorithm to generate a velocity set of each robot in the system. A laser scanner is used to detect obstacles for each robot. Appropriate obstacle avoidance behaviors are generated using sensory information. A behavior‐fusion control scheme combines obstacle avoidance and collision‐free path planning to coordinate multiple robots. The proposed anti‐collision motion planning method for multiple robots prevents collisions with unexpected obstacles and with other robots in the environment. The experimental results show that multiple mobile robots can navigate to targets that are assigned by the cloud server without colliding with other robots, regardless of whether there are unexpected static or dynamic obstacles in the environment.

This paper addresses the complete area coverage problem of a known environment by multiple-robots. Complete area coverage is the problem of moving an end-effector over all available space while avoiding existing obstacles. In such tasks, using multiple robots can increase the efficiency of the area coverage in terms of minimizing the operational time and increase the robustness in the face of robot attrition. Unfortunately, the problem of finding an optimal solution for such an area coverage problem with multiple robots is known to be NP-complete. In this paper we present two approximation heuristics for solving the multi-robot coverage problem. The first solution presented is a direct extension of an efficient single robot area coverage algorithm, based on an exact cellular decomposition. The second algorithm is a greedy approach that divides the area into equal regions and applies an efficient single-robot coverage algorithm to each region. We present experimental results for two algorithms. Results indicate that our approaches provide good coverage distribution between robots and minimize the workload per robot, meanwhile ensuring complete coverage of the area.

Multi-UAV area coverage path planning is a challenging and important task in the field of multi-robots. To achieve efficient and complete coverage in grid-based environments with obstacles and complex boundaries, a multi-UAV area coverage path planning method based on an improved Deep Q-Network (DQN) is proposed in this paper. In the proposed method, a map preprocessing technique based on Depth-First Search (DFS) is introduced to automatically detect and remove unreachable areas. Subsequently, to achieve a reasonable task allocation, the Divide Areas based on Robots’ initial Positions (DARP) algorithm is utilized. In the path planning stage, an enhanced Dueling DQN reinforcement learning architecture is employed by introducing action encoding and prioritized experience replay mechanisms, which improves both training efficiency and policy quality. Moreover, a reward function specifically designed for complete coverage tasks is proposed, effectively reducing redundant visits and mitigating path degradation. Extensive experiments conducted on several benchmark maps show that the proposed method outperforms traditional DQN, Boustrophedon path planning, and Spanning Tree Coverage (STC) methods in terms of coverage rate, redundancy rate, and path length.

The increasing integration of mobile robots in various industries necessitates efficient navigation strategies amidst dynamic environments. Path planning plays a crucial role in guiding mobile robots from their starting points to target destinations, contributing to automation and enhancing human-robot collaboration. This study focuses on devising a tailored path-planning approach for a fleet of mobile robots to navigate through dynamic obstacles and reach designated trajectories efficiently. Leveraging particle swarm optimization (PSO), our methodology optimizes the path while considering real-time environmental changes. We present a simulation-based implementation of the algorithm, where each robot maintains position, velocity, cost, and personal best information to converge towards the global optimal solution. Different obstacles consist of circles, squares, rectangles, and triangles with various colors and five handle-points used. Our findings demonstrate that PSO achieves a global best cost of 5.1017, indicative of the most efficient path, minimizing overall distance traveled.

This paper presents a new approach to the multi-agent coverage path-planning problem. An efficient multi-robot coverage algorithm yields a coverage path for each robot, such that the union of all paths generates an almost full coverage of the terrain and the total coverage time is minimized. The proposed algorithm enables multiple robots with limited sensor capabilities to perform efficient coverage on a shared territory. Each robot is assigned to an exclusive route which enables it to carry out its tasks simultaneously, e.g., cleaning assigned floor area with minimal path overlapping. It is very difficult to cover all free space without visiting some locations more than once, but the occurrence of such events can be minimized with efficient algorithms. The proposed multi-robot coverage strategy directs a number of simple robots to cover an unknown area in a systematic manner. This is based on footprint data left by the randomized path-planning robots previously operated on that area. The developed path-planning algorithm has been applied to a simulated environment and robots to verify its effectiveness and performance in such an application.

As a decision-making layer, path planning technology plays an important role in intelligent driving. Using the A* algorithm or the TEB (timed elastic band) algorithm alone cannot achieve global path optimization and dynamic obstacle avoidance at the same time. This paper adopts a path planning strategy combining A* and TEB algorithms. First, use the A* algorithm to test in a static environment, then use the TEB algorithm to test in a dynamic environment, and finally, conduct experiments in a complex environment with both static and dynamic obstacles, the feasibility and effectiveness of the method for path planning in static maps with dynamic obstacles are verified.

Complex mountainous environments pose significant challenges for dynamic path planning and obstacle avoidance of transport vehicles. In response, this paper presents an innovative path planning approach that combines an enhanced A* algorithm with the artificial potential field (APF) method. Firstly, the heuristic function of the A* algorithm was improved, and path inflection points were optimized to enhance global path-planning efficiency and smoothness. Secondly, a target distance factor was introduced to modify the APF algorithm’s repulsive field function, solving the traditional APF’s target-unreachable problem. The integrated algorithm uses the A*-optimized inflection points as sub-target points for the APF, meeting real-time obstacle avoidance requirements in dynamic environments and conducting secondary path planning to avoid local minima. Impressively, static environment simulations demonstrated the integrated algorithm’s outstanding path-planning capabilities in complex terrains. Moreover, dynamic obstacle avoidance experiments revealed its remarkable ability to not only detect and evade dynamic obstacles but also maintain a safe distance from static ones. The findings highlight that this method significantly boosts path-planning efficiency while ensuring safety and global optimality in dynamic settings. This breakthrough offers crucial theoretical support for enhancing the navigation of mountain transport vehicles in complex, real-world scenarios, potentially improving their operation.

Exploring unknown environments by multiple robots is promising but challenging. The challenges are posed not only by the coordination among multiple robots to improve exploration efficiency, but also by dynamic obstacles that suddenly appear on planned paths. To address those two challenges, this paper proposes a two‐layer architecture where the high‐level layer generates target locations for each robot to explore the unknown environment, while the low‐level layer plans paths in the dynamic environment for each robot. Specifically, in the high‐level design, a novel auction algorithm is proposed, which considers both the distance of robots to target locations and the number of frontiers within the clustering domain of target locations. This approach enables robots to explore different target locations while reducing redundant exploration compared to traditional exploration algorithms. In the low‐level design, a neural network‐based Q‐learning algorithm is employed for path planning to achieve dynamic obstacle avoidance. Robots can dynamically adjust their actions through interaction with the external environment, thus avoid obstacles and reach the target position. To validate our methods, a series of simulation experiments are conducted. The experimental results demonstrate that robots can not only efficiently accomplish exploration tasks in unknown environments, but also achieve effective obstacle avoidance when faced with suddenly appearing dynamic obstacles.

In this paper, a novel DRL algorithm based on a DQN is proposed for multiple mobile robots to find optimized paths. The multiple robots’ states are the inputs of the DQN. The DQN estimates the Q-value of the agents’ actions. After selecting the action with the maximum Q-value, the multiple robots’ actions are calculated and sent to them. Then, the robots will explore the area and detect the obstacles. In the area, there are static obstacles. The robots should detect the static obstacles using a LiDAR sensor. The other moving robots are recognized as dynamic obstacles that need to be avoided. The robots will give feedback on the reward and the robots’ new states. A positive reward will be given when a robot successfully arrives at its goal point. If it is in a free space, zero reward will be given. If the robot collides with a static obstacle or other robots or reaches its start point, it will receive a negative reward. Multiple robots explore safe paths to the goals at the same time, in order to improve learning efficiency. If a robot collides with an obstacle or other robots, it will stop and wait for the other robots to complete their exploration tasks. The episode will end when all robots find safe paths to reach their goals or when all of them have collisions. This collaborative behavior can reduce the risk of collisions between robots, enhance overall efficiency, and help avoid multiple robots attempting to navigate through the same unsafe path simultaneously. Moreover, storage space is used to store the optimal safe paths of all robots. Finally, the multi-robots will learn the policy to find the optimized paths to go to the goal points. The goal of the simulations and experiment is to make multiple robots efficiently and safely move to their goal points.

The path planning is one of the critical technologies for robots to achieve the autonomous operation, enabling to quickly find a safe path in dynamic environments. However, relying on global path planning alone cannot avoid dynamic obstacles, while only using the local path planning may lead to falling into local minima. Therefore, a two-layer robot path planning method suitable for dynamic environments is proposed. This two-layer strategy consists of an efficient global path planning layer and a safe local dynamic obstacle avoidance layer. In the first layer, a parallel sampling and bidirectional guidance rapidly exploring random tree algorithm (PB-RRT) is proposed to search for the global path. To enhance efficiency, the parallel heuristic sampling is introduced to replace the random sampling in bidirectional rapidly exploring random tree (Bi-RRT), and an evaluation function incorporating distance and corner factors is designed to select optimal sampling nodes for adaptive expansion, making the sampling process directional and avoiding over-exploration of space. A bidirectional guidance mechanism further accelerates the merging of the two trees by fully utilizing newly generated node information. Then, a path optimization (PO) method is proposed to improve the length and smoothness of the initial path and obtain the key nodes of the path. In the second layer, the key nodes obtained from the first layer are used as dynamic subtargets, and the safe dynamic window approach (SDWA) is used to achieve the dynamic obstacle avoidance. To further enhance safety, an offset guidance method is proposed to flexibly steer the robot around dynamic obstacles. Extensive experiments show that the average planning time of PB-rapidly exploring random tree (RRT) is reduced by 67.7% compared with Bi-RRT, while the path quality is improved and the path length is reduced by 27.1%. The proposed method also effectively avoids dynamic obstacles in environments with obstacle densities exceeding 60% and achieves the maximum of minimum distance to dynamic obstacles, validating the feasibility and safety of the method.

As different mobile robots may increasingly be used in everyday tasks, it is crucial for path planning to reason about the robots’ physical characteristics. This thesis addresses path planning where robots have to navigate to a destination, either for coverage tasks or perception tasks, which we introduce. For coverage tasks, robots target the destination position, whereas for perception tasks a robot has to reach a point from where a target can be perceived.For complex planning problems with multiple heterogeneous robots, it is essential for planning to run efficiently. This thesis introduces robot-dependent maps, which are map transformations that quickly retrieve information related to the feasibility of coverage and perception tasks. The map transformation depends on properties such as footprint, sensing range and field of view. When dealing with perception tasks, this thesis introduces the concept of perception planning, where paths are calculated not only to minimize motion cost but also to maximize perception quality. In order to find optimal paths for the perception of targets, this work uses an informed heuristic search to determine paths considering both motion and perception. Robot-dependent maps are then used to improve theefficiency of perception planning, by providing dominant heuristics that reduce the number of ray castingoperations and node expansion. The use of robot-dependent maps has a low cost that amortizes over multiple search instances. This thesis also provides a novel technique for multi-robot coverage task planning, with the newrobot-dependent maps used as a pre-processing step. The robot-dependent maps are used to improve theperformance of task allocation when splitting tasks among multiple robots. By using a pre-processing phase, the feasibility of coverage tasks is known for each robot beforehand, and estimates on the cost of each robot executing each task can be quickly computed. This thesis contributes an algorithm to calculate paths for multiple heterogeneous robots that need to perceive multiple regions of interest. Clusters of perception waypoints are determined and heuristically allocated to each robot’s path to minimize overall robot motion and maximize the quality of measures on the regions of interest. Overall, this thesis provides techniques that deal with robot heterogeneity mathematically, modeling the individual characteristics, so that path planning takes into account the perception quality as well. Assuming we have multiple robots with different geometries, ranging from different footprints to differentsensors, the planner accounts for those differences when coordinating the robots. This thesis thus explores algorithms for a planner to dynamically adapt and be able to generate path solutions based on the advantages of each robot. We provide theoretic proofs for some of our contributions and evaluate our algorithms on simulations with a variety of 2D obstacle maps and robots with different physical characteristics.

In this paper, a cellular robot for space trusses is structured so that it can perform tasks such as moving the truss and assembling the truss. There may be some spatial operating mechanisms on the space truss that cause obstacles to the robot’s movement, especially other mobile mechanical devices that are working, which are dynamic obstacles, so a suitable path planning for the robot is needed. In path planning, A-star algorithm has the advantages of efficient searching speed and good optimization effect, but it can’t deal with the path planning problem with dynamic obstacles, so this paper improves Lifelong Planning A-star (LPA-star) algorithm so that the improved algorithm satisfies the dynamic path planning task. Then a three-dimensional truss mathematical model is established, a dynamic obstacle environment is set up, the improved LPA-star algorithm is used for path planning, and the unimproved LPA-star algorithm and the improved A-star algorithm are used to compare with it. The simulation results show that in the environment set up in this paper, the optimal path length of the improved LPA-star algorithm is shortened by about 25% and the algorithm search time is shortened by about 55% compared with the improved A-star algorithm; while the unimproved LPA-star algorithm is unable to accomplish the dynamic path planning task. Therefore, the improved LPA-star algorithm can reduce the robot’s moving distance and time consumption.

Automated vehicles with parallel autonomy show advantages over fully automated vehicles and manual driving. This paper proposes a hierarchical motion planning method that mixes inputs of human drivers and the automated driving systems for automated vehicles in scenarios such as multi-lane roads and multi-intersections with dynamic obstacles. The proposed method comprises a reference path generator in the upper level and a nonlinear model predictive controller with mixed human-vehicle control in the lower level. The path planner considers dynamic obstacles, static obstacles, and human comfort to generate a reference path composed of splines with continuous curvatures in the upper level. In the lower level, the MPC generates a trajectory by tracking the reference path and optimizing the cost function containing inputs of drivers while avoiding both dynamic and static obstacles. The simulation verifies the efficacy and the computational tractability of the proposed method.